The National Highway Traffic Safety Administration (NHTSA), the US transport regulator, has just suggested that a robot may be able to meet the legal definition of a driver. Google’s driverless car project has met obstacles in its need to comply with the Federal Motor Vehicle Safety Standards in order to test their cars on the road. However, the NHTSA has written to the director of Google’s driverless car project to indicate it agrees with their interpretation of a ‘driver’. This means it will interpret ‘driver’ in the context of Google’s driverless car designs, as referring to the driving system and not the person behind the wheel. Anthony Foxx, the US secretary of transport, has promised to lay out new national regulations for these new ‘fully autonomous vehicles’. However this could take a long time to materialise and in the meantime, the NHTSA has allowed Google to apply for an exemption so it can test its driverless cars.
The NHTSA did add that Google should perhaps reconsider its proposals to remove some human controls such as the steering wheel as the vehicle can still achieve autonomy with these controls in place. Google claims their driverless cars are safer if there are no options for humans to override their artificial intelligence system, which bases its decisions on data from sensors, maps and cameras. However, other car manufacturers are promoting a more cautious progression that will retain some control for drivers. Although there remains a ream of legal issues to overcome, if artificial intelligence systems are recognised as having the same legal status as a human driver this could significantly speed up the process of developing driverless cars.